ethers.js/packages/strings/lib/utf8.js

"use strict";
Object.defineProperty(exports, "__esModule", { value: true });
exports.toUtf8CodePoints = exports.toUtf8String = exports._toUtf8String = exports._toEscapedUtf8String = exports.toUtf8Bytes = exports.Utf8ErrorFuncs = exports.Utf8ErrorReason = exports.UnicodeNormalizationForm = void 0;
var bytes_1 = require("@ethersproject/bytes");
var logger_1 = require("@ethersproject/logger");
var _version_1 = require("./_version");
var logger = new logger_1.Logger(_version_1.version);
///////////////////////////////
var UnicodeNormalizationForm;
(function (UnicodeNormalizationForm) {
    UnicodeNormalizationForm["current"] = "";
    UnicodeNormalizationForm["NFC"] = "NFC";
    UnicodeNormalizationForm["NFD"] = "NFD";
    UnicodeNormalizationForm["NFKC"] = "NFKC";
    UnicodeNormalizationForm["NFKD"] = "NFKD";
})(UnicodeNormalizationForm = exports.UnicodeNormalizationForm || (exports.UnicodeNormalizationForm = {}));
;
var Utf8ErrorReason;
(function (Utf8ErrorReason) {
    // A continuation byte was present where there was nothing to continue
    // - offset = the index the codepoint began in
    Utf8ErrorReason["UNEXPECTED_CONTINUE"] = "unexpected continuation byte";
    // An invalid (non-continuation) byte to start a UTF-8 codepoint was found
    // - offset = the index the codepoint began in
    Utf8ErrorReason["BAD_PREFIX"] = "bad codepoint prefix";
    // The string is too short to process the expected codepoint
    // - offset = the index the codepoint began in
    Utf8ErrorReason["OVERRUN"] = "string overrun";
    // A missing continuation byte was expected but not found
    // - offset = the index the continuation byte was expected at
    Utf8ErrorReason["MISSING_CONTINUE"] = "missing continuation byte";
    // The computed code point is outside the range for UTF-8
    // - offset       = start of this codepoint
    // - badCodepoint = the computed codepoint; outside the UTF-8 range
    Utf8ErrorReason["OUT_OF_RANGE"] = "out of UTF-8 range";
    // UTF-8 strings may not contain UTF-16 surrogate pairs
    // - offset       = start of this codepoint
    // - badCodepoint = the computed codepoint; inside the UTF-16 surrogate range
    Utf8ErrorReason["UTF16_SURROGATE"] = "UTF-16 surrogate";
    // The string is an overlong representation
    // - offset       = start of this codepoint
    // - badCodepoint = the computed codepoint; already bounds checked
    Utf8ErrorReason["OVERLONG"] = "overlong representation";
})(Utf8ErrorReason = exports.Utf8ErrorReason || (exports.Utf8ErrorReason = {}));
;
function errorFunc(reason, offset, bytes, output, badCodepoint) {
    return logger.throwArgumentError("invalid codepoint at offset " + offset + "; " + reason, "bytes", bytes);
}
function ignoreFunc(reason, offset, bytes, output, badCodepoint) {
    // If there is an invalid prefix (including stray continuation), skip any additional continuation bytes
    if (reason === Utf8ErrorReason.BAD_PREFIX || reason === Utf8ErrorReason.UNEXPECTED_CONTINUE) {
        var i = 0;
        for (var o = offset + 1; o < bytes.length; o++) {
            if (bytes[o] >> 6 !== 0x02) {
                break;
            }
            i++;
        }
        return i;
    }
    // This byte runs us past the end of the string, so just jump to the end
    // (but the first byte was read already read and therefore skipped)
    if (reason === Utf8ErrorReason.OVERRUN) {
        return bytes.length - offset - 1;
    }
    // Nothing to skip
    return 0;
}
function replaceFunc(reason, offset, bytes, output, badCodepoint) {
    // Overlong representations are otherwise "valid" code points; just non-deistingtished
    if (reason === Utf8ErrorReason.OVERLONG) {
        output.push(badCodepoint);
        return 0;
    }
    // Put the replacement character into the output
    output.push(0xfffd);
    // Otherwise, process as if ignoring errors
    return ignoreFunc(reason, offset, bytes, output, badCodepoint);
}
// Common error handing strategies
exports.Utf8ErrorFuncs = Object.freeze({
    error: errorFunc,
    ignore: ignoreFunc,
    replace: replaceFunc
});
// http://stackoverflow.com/questions/13356493/decode-utf-8-with-javascript#13691499
function getUtf8CodePoints(bytes, onError) {
    if (onError == null) {
        onError = exports.Utf8ErrorFuncs.error;
    }
    bytes = (0, bytes_1.arrayify)(bytes);
    var result = [];
    var i = 0;
    // Invalid bytes are ignored
    while (i < bytes.length) {
        var c = bytes[i++];
        // 0xxx xxxx
        if (c >> 7 === 0) {
            result.push(c);
            continue;
        }
        // Multibyte; how many bytes left for this character?
        var extraLength = null;
        var overlongMask = null;
        // 110x xxxx 10xx xxxx
        if ((c & 0xe0) === 0xc0) {
            extraLength = 1;
            overlongMask = 0x7f;
            // 1110 xxxx 10xx xxxx 10xx xxxx
        }
        else if ((c & 0xf0) === 0xe0) {
            extraLength = 2;
            overlongMask = 0x7ff;
            // 1111 0xxx 10xx xxxx 10xx xxxx 10xx xxxx
        }
        else if ((c & 0xf8) === 0xf0) {
            extraLength = 3;
            overlongMask = 0xffff;
        }
        else {
            if ((c & 0xc0) === 0x80) {
                i += onError(Utf8ErrorReason.UNEXPECTED_CONTINUE, i - 1, bytes, result);
            }
            else {
                i += onError(Utf8ErrorReason.BAD_PREFIX, i - 1, bytes, result);
            }
            continue;
        }
        // Do we have enough bytes in our data?
        if (i - 1 + extraLength >= bytes.length) {
            i += onError(Utf8ErrorReason.OVERRUN, i - 1, bytes, result);
            continue;
        }
        // Remove the length prefix from the char
        var res = c & ((1 << (8 - extraLength - 1)) - 1);
        for (var j = 0; j < extraLength; j++) {
            var nextChar = bytes[i];
            // Invalid continuation byte
            if ((nextChar & 0xc0) != 0x80) {
                i += onError(Utf8ErrorReason.MISSING_CONTINUE, i, bytes, result);
                res = null;
                break;
            }
            ;
            res = (res << 6) | (nextChar & 0x3f);
            i++;
        }
        // See above loop for invalid continuation byte
        if (res === null) {
            continue;
        }
        // Maximum code point
        if (res > 0x10ffff) {
            i += onError(Utf8ErrorReason.OUT_OF_RANGE, i - 1 - extraLength, bytes, result, res);
            continue;
        }
        // Reserved for UTF-16 surrogate halves
        if (res >= 0xd800 && res <= 0xdfff) {
            i += onError(Utf8ErrorReason.UTF16_SURROGATE, i - 1 - extraLength, bytes, result, res);
            continue;
        }
        // Check for overlong sequences (more bytes than needed)
        if (res <= overlongMask) {
            i += onError(Utf8ErrorReason.OVERLONG, i - 1 - extraLength, bytes, result, res);
            continue;
        }
        result.push(res);
    }
    return result;
}
// http://stackoverflow.com/questions/18729405/how-to-convert-utf8-string-to-byte-array
function toUtf8Bytes(str, form) {
    if (form === void 0) { form = UnicodeNormalizationForm.current; }
    if (form != UnicodeNormalizationForm.current) {
        logger.checkNormalize();
        str = str.normalize(form);
    }
    var result = [];
    for (var i = 0; i < str.length; i++) {
        var c = str.charCodeAt(i);
        if (c < 0x80) {
            result.push(c);
        }
        else if (c < 0x800) {
            result.push((c >> 6) | 0xc0);
            result.push((c & 0x3f) | 0x80);
        }
        else if ((c & 0xfc00) == 0xd800) {
            i++;
            var c2 = str.charCodeAt(i);
            if (i >= str.length || (c2 & 0xfc00) !== 0xdc00) {
                throw new Error("invalid utf-8 string");
            }
            // Surrogate Pair
            var pair = 0x10000 + ((c & 0x03ff) << 10) + (c2 & 0x03ff);
            result.push((pair >> 18) | 0xf0);
            result.push(((pair >> 12) & 0x3f) | 0x80);
            result.push(((pair >> 6) & 0x3f) | 0x80);
            result.push((pair & 0x3f) | 0x80);
        }
        else {
            result.push((c >> 12) | 0xe0);
            result.push(((c >> 6) & 0x3f) | 0x80);
            result.push((c & 0x3f) | 0x80);
        }
    }
    return (0, bytes_1.arrayify)(result);
}
exports.toUtf8Bytes = toUtf8Bytes;
;
function escapeChar(value) {
    var hex = ("0000" + value.toString(16));
    return "\\u" + hex.substring(hex.length - 4);
}
function _toEscapedUtf8String(bytes, onError) {
    return '"' + getUtf8CodePoints(bytes, onError).map(function (codePoint) {
        if (codePoint < 256) {
            switch (codePoint) {
                case 8: return "\\b";
                case 9: return "\\t";
                case 10: return "\\n";
                case 13: return "\\r";
                case 34: return "\\\"";
                case 92: return "\\\\";
            }
            if (codePoint >= 32 && codePoint < 127) {
                return String.fromCharCode(codePoint);
            }
        }
        if (codePoint <= 0xffff) {
            return escapeChar(codePoint);
        }
        codePoint -= 0x10000;
        return escapeChar(((codePoint >> 10) & 0x3ff) + 0xd800) + escapeChar((codePoint & 0x3ff) + 0xdc00);
    }).join("") + '"';
}
exports._toEscapedUtf8String = _toEscapedUtf8String;
function _toUtf8String(codePoints) {
    return codePoints.map(function (codePoint) {
        if (codePoint <= 0xffff) {
            return String.fromCharCode(codePoint);
        }
        codePoint -= 0x10000;
        return String.fromCharCode((((codePoint >> 10) & 0x3ff) + 0xd800), ((codePoint & 0x3ff) + 0xdc00));
    }).join("");
}
exports._toUtf8String = _toUtf8String;
function toUtf8String(bytes, onError) {
    return _toUtf8String(getUtf8CodePoints(bytes, onError));
}
exports.toUtf8String = toUtf8String;
function toUtf8CodePoints(str, form) {
    if (form === void 0) { form = UnicodeNormalizationForm.current; }
    return getUtf8CodePoints(toUtf8Bytes(str, form));
}
exports.toUtf8CodePoints = toUtf8CodePoints;
//# sourceMappingURL=utf8.js.map